JP3216503B2 - Multiplexing device and multiplexing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multiplexing apparatus for performing multiplexing for encoding and transmitting or storing multimedia data such as digitized audio, video, and other data. It is applied to a channel program multiplexing device. Specific application examples include digital broadcasting systems using satellites, terrestrial waves, cables, and optical fibers, video on demand (VOD), video conference systems, and the like.

[0002]

[Prior art]

Conventional example 1. FIG. 11 shows, for example, Hattori, Tanaka, Asano, Asai,
Sakado: "Satellite-based broadcast quality codec", Mitsubishi Electric Technical Report, Vol. 67, no. 7, pp. 33-38 (19
FIG. 93 is a configuration diagram of a multimedia multiplexing device shown in 93). In FIG. 11, reference numeral 110 denotes a video encoding unit that performs information source encoding on the video signal 101 to generate a video encoded bit sequence 104, and reference numeral 111 denotes audio encoding that performs information source encoding on the audio signal 102 to generate an audio encoded bit sequence 105. And 112, a data encoding unit that performs various processes on the data signal 103 to generate a data encoded bit sequence 106, and 100 denotes a video encoded bit sequence 10
4 and the audio encoded bit sequence 105 and the data encoded bit sequence 106 are multiplexed into one multiplexed bit sequence 107.
Is a conventional media multiplexing unit.

Next, the operation of the conventional multimedia multiplexing device will be described. The video signal 101, the audio signal 102, and the data signal 103 are respectively converted into a video encoder 110, an audio encoder 111, and a data encoder 1
12, the source is coded and the video coded bit sequence 104
An audio encoded bit string 105 and a data encoded bit string 106 are generated. In the media multiplexing unit 100,
The video encoding bit sequence 111, the audio encoding bit sequence 112, and the data encoding bit sequence 113 are multiplexed into one multiplexed bit sequence 107.

[0004] Multiplexing is performed in units of predetermined multiplexed frames. FIG. 12 shows a conventional media multiplexing unit 10.
FIG. 4 is an explanatory diagram relating to a multiplexing method at 0. In the figure, reference numeral 120 denotes a multiplexed frame for transmitting a multiplexed bit sequence, 121 denotes a synchronization bit for ensuring multiplexed frame synchronization, and 122 denotes a video coded bit sequence 104. Reference numeral 123 denotes an audio slot in which the audio coded bit string 105 is stored, and reference numeral 124 denotes a data slot in which the data coded bit string 106 is stored.

The multiplexing of the coded data of each medium is performed in units of multiplexed frames having a predetermined repetition frequency.
4 is time-division multiplexed by containing the audio coded bits 105 in the audio slot 123 and the data coded bit string 106 in the data slot 124. In order to ensure synchronization of the multiplexed frame 120 on the receiving side, a synchronization bit 121 of a unique word is arranged at the first bit of the multiplexed frame 120.

Conventional example 2. FIG. 13 shows another conventional example described in JP-A-4-219041. In FIG. 13, 125a to 125n are buffers for storing packets input from the input ports 126a to 126n, 127a to 127n are counters for counting the number of waiting packets stored in the buffers 125a to 125n, and 128 is a counter for counting the number of waiting packets. Counter 127a-
127n, and a limiting polling unit that determines a transmission port from the number of waiting packets and an offset value.
Reference numeral 9 denotes a queue length monitoring unit which is connected to the counters 127a to 127n and determines a transmission port from the number of waiting packets and the offset value. Reference numeral 130 denotes a counter 127 corresponding to the buffer 125 in which media with the most severe discard request are stored. By comparing the count value with the threshold value of the number of queued packets for the media whose discard request is the strictest, the limit switching unit that switches between the limiting polling unit and the queue length monitoring unit,
A switch that selects and outputs one of n based on the control from the limiting expression polling unit 128 or the queue length monitoring unit 129. A multiplexing control device 132 is configured by the above-mentioned restriction type polling unit 128, queue length monitoring unit 129, and restriction switching unit 130. In such a multimedia packet multiplexing apparatus, first, packets input from the input ports 126a to 126n are temporarily stored in the buffers 125a to 125n, and at that time, the corresponding counter increases the value by one.

[0007] The limit switching unit 130 includes a threshold value for the number of waiting packets for a medium with the strictest discard request and a buffer 125 for storing the media with the strictest discard request.
Is compared with the value of the counter 127, and if the number of waiting packets is equal to or smaller than the threshold value, the restricted polling unit 128
When the value is equal to or larger than the threshold, an instruction to use the queue length monitoring unit 129 is issued.

The limiting polling unit 128 sequentially selects each of the buffers 125a to 125n, inputs the buffer to the switch 131, and sends out one packet at a time. Here, the limiting-type polling unit 128 has an offset value corresponding to each of the input ports 126a to 126n that indicates a smaller value as the delay request is higher, and the value of each of the counters 127a to 127n is equal to or less than the respective offset value. In this case, the number of waiting packets is regarded as 0, and the port is skipped. This allows
Media packets that are strictly delayed are transmitted with priority.

Further, the queue length monitoring unit 129 regards a medium having a higher discard request as having a lower offset value with respect to the number of waiting packets, and regards a value obtained by subtracting this offset value from the number of waiting packets as a logical number of waiting packets. Select the port with the largest number of logical waiting packets. As a result, packets of media that are strictly discarded are transmitted with priority.

After each port is selected, one packet is sent out as a multiplexed output, and the value of the counter of the selected port is decreased by one. With the above operation, priority control of packets of media strictly discarded and priority control of media strictly delayed are independently performed.

[0011]

In the multimedia multiplexing apparatus shown in the above-mentioned prior art example 1, the coded data of each medium is time-division multiplexed at a predetermined fixed ratio, and each medium is multiplexed. The encoded data will be multiplexed at a fixed bit rate.

Therefore, when variable rate (bit rate) encoding is performed such that the bit rate changes with time, a slot having a width corresponding to the maximum bit rate of the encoded data of each medium in a transmission frame. Unless data is secured for all media, complete data transmission is not possible. When such a slot is prepared, the size of the multiplexed frame becomes large, so that it takes a long time to secure synchronization of the multiplexed frame on the receiving side or when the encoded data does not reach the maximum bit rate. Means that dummy data is transmitted to the rest of the corresponding slot, so that the transmission efficiency is reduced.In addition, in the multiplexed bit stream after multiplexing, only a fixed transmission rate is given to each medium, so that transmission is desired. There is a problem that it is not possible to flexibly adapt to the increase or decrease of the encoded data, and a transmission delay due to multiplexing occurs.

Especially when the number of media to be multiplexed is large, the size of the multiplexed frame becomes large, so that a program (program) composed of video, audio, data, etc.
In a multi-channel broadcasting service in which a plurality of channels are collected and multiplexed into one multiplexed frame and delivered using a satellite or a coaxial cable, the reduction in multiplexing efficiency and transmission delay become remarkable.

Further, since video, audio, and data are multiplexed by storing them at predetermined fixed positions in a multiplexed frame, in addition to the above-described change in the bit rate of each medium, However, it has been difficult to flexibly respond to changes in the content of transmission data such as the presence or absence of media and efficiently multiplex.

In the multimedia packet multiplexing apparatus shown in the conventional example 2, since the number of packets stored in the buffer is used as a criterion for selecting a transmission packet, for example, the number of packets transmitted from each input port depends on the type of media. If the size of incoming packets is different, the multiplexing controller cannot properly detect the buffer capacity vacancy only by the number of packets, so the input packets must be input to perform appropriate control without buffer overflow. 1
The capacity of each buffer must be set individually in consideration of the size per unit (particularly the maximum amount), and the amount of data generated per unit time varies due to variable bit rate encoding as described above. In this case, there is a problem in that it is not possible to perform feedback control on the operation of variable bit rate coding in accordance with the amount of data, not the number of data being buffered, and to suppress transmission delay due to multiplexing.

The present invention has been made to solve the above problems, and flexibly and appropriately responds to changes in the transmission speed of individual media to be multiplexed, the presence or absence of individual media, and the like. Multiplex transmission can be performed efficiently, and the delay time associated with multiplexing can be suppressed.

[0017]

A multiplexing apparatus according to the present invention comprises: Multiplexing control means for selecting from coded data stored in the multiplexing buffer means.

Further, the multiplexing control means is preferentially selected as a multiplexed output from the coded data stored in the multiplexing buffer means having the largest storage amount among the stored data amounts of the multiplexing buffer means. It is something to do.

Alternatively, the multiplexing control means has weighting means for weighting the amount of data stored in each of the multiplexing buffer means, and selects a multiplexed output based on the amount of stored data weighted by the weighting means. It is like that.

In addition, an invalid data generating means for generating invalid data is provided, and the multiplexing control means is provided with the invalid data generating means when none of the stored data amounts of the multiplexing buffer means is less than a predetermined amount. Is selected as multiplexed output.

Further, there are provided transmission control data generating means for generating transmission control data, and interrupt timer means for measuring an interval of transmission of the transmission control data, and an output of the transmission control data generating means based on a result of the timer by the interrupt timer means. Is selected as the multiplexed output.

If the amount of data stored in each of the multiplexing buffer means is less than a predetermined amount, the transmission control data output from the transmission control data generating means is selected as a multiplexed output, and the interrupt timer is selected. The means is set to a timing start state.

Further, the multiplexing buffer means includes control signal output means for outputting a control signal of a generated code amount of the coded data to be inputted, based on the accumulated data amount of each of the multiplexing buffer means. is there.

Further, the multiplexing control means includes a control signal output means for outputting a control signal of a generated code amount of the coded data to be inputted based on the amount of data stored in each of the multiplexing buffer means. .

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION

Embodiment 1 of the Invention Hereinafter, an embodiment of a multiplexing device of the present invention will be described with reference to the drawings. In FIG. 1, 10a
10 to 10 n are n encoding units for encoding various media information 20 a to 20 n such as video, audio, and data;
Reference numerals 11a to 11n denote multiplexing buffers as n multiplexing buffer means for storing coded data 21a to 21n output from the coding units 10a to 10n, and 12 denotes a multiplexing buffer stored in the multiplexing buffers 11a to 11n. A multiplexing unit as multiplexing means for multiplexing the coded data and outputting the multiplexed data as a multiplexed stream 24.
The multiplexing unit 13 multiplexes the coded data to be output based on the accumulated data amount of the coded data 21a to 21n in the plurality of multiplexing buffers 11a to 11n.
11n is a multiplex control unit as multiplex control means for selecting from encoded data stored in 11n.

Next, the operation will be described. Various types of media information 20a to 20n such as video, audio, and data are individually taken into the encoding units 10a to 10n corresponding to various applications, and are subjected to information source encoding processing to produce encoded data 21a to 21n. Generated.
The individual coded data 21a to 21n are input to the multiplexing buffers 11a to 11n and accumulated until multiplexed.

In the multiplexing section 12, each multiplexing buffer 11a
11n are multiplexed and output as a multiplexed stream 24. Here, each of the multiplexing buffers 11a to 11n includes an encoding unit 10a to
Each time the encoded data 21a to 21n from 10n is accumulated or the data accumulated in the multiplexing unit 12 is output,
Buffer storage amounts 23a to 23 indicating remaining data storage amounts
Output n.

The multiplexing control unit 13 controls the buffer storage amounts 23a to 23d input from the multiplexing buffers 11a to 11n.
One coded data to be multiplexed is selected based on 23n, and a multiplexing selection signal 25 is output.

The multiplexing unit 12 receives the encoded data 2
The encoded data indicated by the multiplexing selection signal 25 among 2a to 22n is multiplexed one after another and output as a multiplexed stream 24.

Here, an example of a method of selecting encoded data performed in the multiplex control unit 13 will be described with reference to FIG. In FIG. 2, reference numeral 27 denotes a buffer storage amount 2 inputted from each of the multiplexing buffers 11a to 11n.
3a to 23n, and selects the coded data stored in the multiplexing buffer having the largest storage amount, and a multiplexing selection signal 2 indicating its number (hereinafter referred to as a stream number).
5 is a buffer accumulation amount comparison unit that outputs 5.

Next, the operation will be described. The buffer storage amount comparing section 27 compares the buffer storage amounts 23a to 23n from the multiplexing buffers 11a to 11n, selects the one having the maximum storage amount, and sets the corresponding stream number to a multiplexing selection signal. Output as 25.

For example, at a certain point in time, the buffer storage amount 23a
, The stream number a is output as the multiplex selection signal 25. Thereby, the multiplexing buffer 11a having the largest buffer storage amount is selected, and the coded data 22a from the multiplexing buffer 11a is
Are multiplexed and output by the multiplexing buffer 11a.
Data storage amount decreases.

The operation from selection of coded data to be multiplexed to multiplexing will be described in more detail with reference to FIG. In FIG.
Assuming that the number of coded streams to be multiplexed is three, the change with time of the buffer accumulation amount of the multiplexing buffers 11a, 11b, and 11c is shown. That is, the time t
Buffer storage amount 23a of the multiplexing buffer 11a at
Are indicated by Sa (t), and similarly, the buffer accumulation amounts 23b and 23c of the multiplexing buffers 11b and 11c are indicated by Sb (t) and Sc (t), respectively.

At times t1, t2,. . Indicates the multiplexing timing. Here, for simplicity, it is assumed that the multiplexing selection signal 25 is selected at fixed time intervals and the encoded data of the selected multiplexing selection signal 25 is multiplexed. I do.

Here, each of the encoded data 2
It is assumed that the information generation speed of 1a, 21b, 21c is constant, and that the sum of the information generation speeds of the encoded data 21a, 21b, 21c is equal to the transmission speed of the multiplexed stream 24.

At time t1, the buffer storage amount comparison unit 30 stores the storage amounts 23a, 23a,
23b and 23c are compared. At time t1, as shown in FIG. 3, Sb (t1)> Sa (t1)> Sc (t
1), the stream number b is selected, and b is output as the multiplexing selection signal 25.

The multiplexing unit 12 reads out m bits from the multiplexing buffer 11 b in accordance with the multiplexing selection signal 25 and multiplexes the bits into a multiplexed stream 24. Here, for simplicity, it is assumed that the encoded data 22b is instantaneously read from the multiplexing buffer b11b. Therefore, at the next time point when the storage amounts are compared at time t1, the storage amount Sb (t1) of the multiplexing buffer 11b decreases by m bits.

Actually, the m bits are output as the multiplexed stream 24 from time t1 to time t2 of the next multiplexing timing, and the transmission speed of the multiplexed stream 24 is constant (CBR: Constant Bit).
Rate), the transmission rate at that time is: transmission rate [bit / sec] = m [bit] / (t2−t
1) Given in [seconds].

Next, at time t2, as shown in FIG.
Since (t2)> Sb (t2)> Sc (t2), the stream number a is selected, and a is output as the multiplexing selection signal 25. Thereafter, similarly, at times t3, t4, t
5, the buffer accumulation amount is compared at t6, and b, c, b, and a are sequentially output as the multiplexing selection signal 25.
In accordance with this, encoded data is read from each multiplexing buffer.

When the multiplexing buffer storage amounts are the same, a predetermined priority (for example, a stream having a smaller stream number to a higher priority, or a stream having a lower coded stream information generation speed has a higher priority). ), And selects a coded stream number, and outputs a multiplexing selection signal 25.

FIG. 4 shows an example of encoded data 21a to 21n output from the encoding units 10a to 10n, where 35 is a fixed length packet composed of a fixed length (m bits), and 36 is a packet. A packet header 37 includes information for performing identification and the like, and a payload 37 includes encoded data and control information.

In the encoding units 10a to 10n, the respective encoded data are accommodated in the payload 37 of the fixed-length packet 35, and are multiplexed in the multiplexing buffer 1 in the form of the fixed-length packet 35.
1a to 11n. Packet header 3 at that time
6 includes a packet identification number unique to each medium so that the contents of the payload 37 can be distinguished.

The multiplexing unit 12 performs multiplexing in units of m-bit fixed-length packets 35 to generate a multiplexed stream 24. On the receiving side, the multiplexed stream 24 is separated into m-bit fixed packet length units, and each fixed-length packet 3
5 are separated from each other by the packet header 36.

By the above operation, the coded data to be multiplexed is selected according to the amount of data stored in the multiplexing buffer and multiplexed and output, so that it is not necessary to determine a fixed multiplexing pattern in advance and efficient multiplexing transmission Can be performed.

Since multiplexing is performed by selecting a buffer having a large storage capacity, the required capacity of the multiplexing buffer can be minimized and the delay time associated with multiplexing can be reduced.

Further, the selection of the coded data to be multiplexed is
Since the control is performed based on the accumulation amount of the multiplexing buffer, it is possible to perform control reflecting the data accumulation amount itself in the multiplexing buffer.

For example, see the embodiment shown in FIG.
Assuming that one output packet has m bits, each multiplexing buffer
The number of packets stored in the router is 0 or 1, and FIG.
Similar operation is performed by comparing the number of packets as before.
Which multiplexing buffer should read the data
I can't judge. In order to make a judgment by comparing the number of packets, the capacity of the multiplexing buffer must be further increased, which increases the data transmission delay accompanying the multiplexing.

The bi accumulated amount of multiplexed buffer as above
If the control is performed based on the amount of data stored in units of bits and the amount of storage in units of bits, it is possible to perform fine multiplexing control that reflects not the number of packets in the multiplexing buffer but the amount of data stored in the multiplexing buffer itself. And transmission delay can be suppressed as much as possible.

In the above embodiment, the transmission speed (bit rate) of the multiplexed stream 24 is constant (CBR: C
Although the present invention has been described as an example of a VBR (Variable Bit Rate), the transmission rate of actual data in the multiplexed stream 24 is variable.
e Bit Rate).

In the case of this VBR, the amount of encoded data generated by the encoding section is usually variable, and transmission using VBR is suitable for transmitting data in which the amount of generated data changes. At this time, if the multiplexing timing is fixed, the number of multiplexed bits is variable. That is, when the multiplexing timing is set at a constant interval as in the above-described embodiment described with reference to FIG. 3, the number of bits read for multiplexing is changed.

In other words, since the amount of encoded data generated from the encoding unit is variable, the amount of accumulation in each multiplexing buffer does not increase in a fixed manner with respect to the time axis. However, when data is generated, the storage amount may increase suddenly.However, when reading out a fixed number of bits, the capacity of the multiplexing buffer must be increased, and the stream is selected many times, and the interval for multiple times is selected. Therefore, the transmission delay of the stream and other streams is increased. Therefore, the number of bits read at a time is increased (that is, the number of multiplexed bits is increased) to suppress the transmission delay.

In the present invention, the coded data to be multiplexed is selected in view of the storage amount itself, so that it is possible to grasp how the storage amount increases, and the number of bits read at a time may be determined according to each storage amount. Such an operation is possible because the accumulated amount itself is used as a reference.

When multiplexing is performed with m bits fixed, multiplexing timing becomes possible. That is, in the above-described embodiment described with reference to FIG. 3, the multiplexing timing is set at regular intervals, and the coded data to be multiplexed is selected by comparing the accumulated amount at that time. Although the structure can be relatively simplified, in the case of VBR,
Problem arises. Therefore, when any one of the multiplexing buffers reaches a predetermined storage amount, the coded data is selected and multiplexed. In this case, the multiplexing timing becomes variable. It is possible to perform such an operation at a fine level by grasping the accumulated amount.

Further, as shown in FIG.
When packetized by a to 10n, the packets can be multiplexed in the unit and the configuration of the multiplexing unit can be simplified. On the other hand, the multiplexing buffer or the multiplexing unit attaches a header indicating each stream to each encoded data. By doing so, the delimiter of the stored data read from the multiplexing buffer can be freely set, so that it is possible to cope with the case of VBR.

For example, it is possible to determine a lower limit threshold value in the multiplexing buffer accumulation amount in FIG. 3 and read out the selected stream up to the lower limit threshold value as a multiplexed output at a time. If this threshold is changed for each buffer, each stream (media)
Can be transmitted.

In the above description, each multiplexing buffer 1
Although the output of the storage amounts 23a to 23n from 1a to 11n has been described, the remaining buffer amount may be output instead of the storage amount.
-11n, and a counter for counting the amount of data stored in each of the multiplexing buffers 11a-11n.
A signal indicating the output data amount may be sequentially transmitted, and the accumulated amount of each buffer may be grasped and determined by the counter of the multiplexing control unit 13. In short, the data accumulated amount of each of the multiplexed buffers 11a to 11n may be grasped. .

Embodiment 2 of the Invention In the first embodiment, the multiplexed buffers 11a to 11n are compared equally to select the coded data to be multiplexed. However, coded data of a specific medium may be preferentially multiplexed with a low delay time. In this case, the buffer storage amount for the other coded data may be input to the multiplexing control unit 13 after being reduced from the actual value.

When the overall priority is given, each accumulation amount is weighted, and the selection may be made based on the weighted accumulation amount. An embodiment thereof will be described with reference to FIG. In FIG. 5, reference numeral 50 denotes a weighting coefficient table;
Is a weighting section, and 52 is a comparison section, and the same reference numerals as those in FIG. 1 indicate the same or corresponding parts.

Next, the operation will be described. Multiplex control unit 1
3 are stored in the weighting factors 53a to 53n input from the weighting factor table 50.
The weighting unit 51 performs weighting according to a to 53n.

The values of the weighting coefficients 53a to 53n are determined, for example, by the allowable delay time of each medium (stream) for multiplexing, and a large weight is applied to a medium having a short allowable delay time.

As a result, the buffer storage amount 5 after weighting
When the comparison of 4a to 54n is performed by the comparing unit 52, the medium having the shortest allowable delay time is preferentially selected, and the multiplexing selection signal 25 is output. Generally, audio media has a smaller buffer size than video media, and the allowable delay time is shorter than that of video media. Therefore, a large weight is assigned to the audio media, and control is performed so that the delay time due to multiplexing is reduced. In this way, the priority control is performed by performing fine weighting using the value of the accumulation amount.

Embodiment 3 of the Invention FIG. 6 is a configuration diagram showing another embodiment of the present invention. In FIG. 6, 30
Denotes a multiplexing unit, 31 denotes an invalid data generating unit, and the same reference numerals as those in FIG. 1 denote the same or corresponding parts.

Next, the operation will be described. When the sum of the information generation speeds of the encoded data 22a to 22n is smaller than the transmission speed of the multiplexed stream 33, the invalid data generation unit 31
More invalid (null) data 32 is generated and multiplexed by the multiplexing unit 30 onto the multiplexed stream 33,
3, the transmission speed is matched.

That is, when the buffer storage amounts 23a, 23b, 23c of the multiplexing buffers 11a, 11b, 11c at the multiplexing timing as shown in FIG. Is output, and multiplexing is performed. Further, as the invalid data 32, the bit values of the data are all “1” or “0”.

Embodiment 4 of the Invention FIG. 7 is a configuration diagram showing another embodiment of the present invention. In FIG. 7, 40
Is a transmission control data generator as transmission control data generator, 41 is an interrupt timer as an interrupt timer, and 42 is a multiplexing unit. The same reference numerals in FIG. 1 denote the same or corresponding parts.

Next, the operation will be described. The transmission control data generation unit 40 performs transmission control that needs to be periodically transmitted, such as transmission clock synchronization information necessary for ensuring transmission / reception clock synchronization, control information for media information included in the multiplexed stream 45, and auxiliary information. Data 43 is generated.

Multiplexed stream 4 of transmission control data 43
The multiplexing into 5 is performed by an interrupt signal 44 periodically generated from the interrupt timer unit 41. That is, when the interrupt signal 44 occurs, the multiplexing unit 42 ignores the coded data selection by the multiplexing selection signal 25, selects the control data 43, and performs multiplexing on the multiplexed stream 45. When the interrupt signal 44 is generated, the timer of the interrupt timer unit 41 is set again, and the interrupt signal 44 is generated again after a predetermined time has elapsed. It should be noted that the amount of the coded data that has been ignored is not reduced, so that it is selected at the next multiplexing timing.

It should be noted that the transmission control data generating means and the interrupt timer means shown in this embodiment may be combined with the invalid data generating means shown in the third embodiment.

FIG. 8 is an explanatory diagram for explaining the operation in such an embodiment. In FIG. 8, for simplicity, the number of encoded streams is three, and the information generation speed and the multiplexing cycle of each medium are constant, as in FIG.

Times t0, t1, which are multiplexing timings
At t2, each media (multiplexing buffer 10a,
10b, 10c) are compared, and
Among them, the one having the largest storage amount is selected, and a multiplexed stream is generated as shown in FIG. However, the buffer storage amounts Sa (t3), Sb (t3),
If Sc (t3) is less than the number of bits required for multiplexing, the multiplexing control unit 13 outputs a multiplexing selection signal 2 indicating invalidity.
5 is output, and the invalid data 42 is multiplexed.

In FIG. 8, at time t3, the buffer accumulation amounts Sa (t3), Sb (t3), Sc (t
Since 3) is less than m bits required for multiplexing, the multiplexing control unit 13 outputs a multiplexing selection signal 25 indicating invalidity, and the invalid data 42 is multiplexed. Next, at time t5, the stream number c is selected by comparing the accumulated amount of the multiplexing buffer. The generation of the interrupt signal 44 causes the multiplexing unit 42 to select the transmission control data 43 and perform the multiplexing.

Embodiment 5 of the Invention Also, the buffer accumulation amounts 23a, 23a, 2b of the multiplexing buffers 11a, 11b, 11c.
When each of 3b and 23c is less than m bits,
The multiplexing selection signal 25 input to the interrupt timer unit 41
Therefore, if the interrupt signal 44 is generated regardless of the timer value, the transmission control data 43 can be multiplexed instead of the invalid data 42, and the multiplexing efficiency can be improved. After the transmission control data 43 is multiplexed, the timer of the interrupt timer unit 41 is set again.

Embodiment 6 of the Invention FIG. 9 is a configuration diagram showing another embodiment of the present invention. In FIG. 9, FIG.
The same reference numerals indicate the same or corresponding ones.

Next, the operation will be described. In FIG. 9, in each of the multiplexing buffers 11a to 11n, the current buffer storage amount is subtracted from the capacity of each of the multiplexing buffers to obtain buffer remaining amounts 26a to 26n. Control signal output means for outputting the control signal to each of the encoding units 10a to 10n is provided.

In the encoding unit 10a, the remaining buffer 26a
The information generation speed of the encoded data 21a is controlled according to the magnitude of. That is, when the remaining buffer amount 26a is small, the information generation amount of the encoded data 21a is reduced, and when the buffer remaining amount 26a is large, the information generation amount of the encoded data 21a is increased.

As a result, it is possible to prevent the accumulated amount of the encoded data 21a from overflowing the capacity of the multiplexing buffer 11a, and to reduce the multiplexing delay by controlling the accumulated amount of the multiplexing buffer 11a. it can. Here, the encoding unit 10a has been described as an example, but the same operation is performed on the other encoding units 10b to 10n.

Embodiment 7 of the Invention FIG. 10 is a configuration diagram showing another embodiment of the present invention. In FIG.
80a to 80n are encoding units, 81 is a multiplex control unit,
The same reference numerals as those in FIG. 1 indicate the same or corresponding components.

Next, the operation will be described. 10, the current buffer storage amounts 23a to 23n are input to the multiplexing control unit 81 from the multiplexing buffers 11a to 11n. The multiplexing control unit 81 selects the coded stream to be multiplexed according to the magnitude of the buffer storage amounts 23a to 23n as described in the first embodiment of the invention, and outputs the multiplexed signal 25 to the multiplexing unit 12.

The multiplexing control unit 81 obtains the total storage amount from the buffer storage amounts 23a to 23n, and according to the magnitude thereof, encodes control signals 82a to 82a as control signals for the generated code amount of the encoded data. 82n to each encoding unit 8
Control signal output means for outputting signals to 0a to 80n is provided.

For example, when the total storage amount is large, it indicates that a multiplexing delay has occurred in the multiplexing apparatus as a whole, and thus control is performed so as to suppress the entire coded data generation amount. Conversely, when the total storage amount is small and a large amount of invalid data is generated, control is performed so as to increase the entire encoded data generation amount in order to increase the multiplexing efficiency.

As a result, the multiplexing delay time of the entire multiplexing apparatus can be reduced and the multiplexing efficiency can be improved. Further, since the amount of information generated by the encoding unit is controlled by the remaining amount of the multiplexing buffer, it is possible to adaptively follow the transmission speed of the multiplexed stream, and always efficient multiplexing is performed using the transmission band effectively.

[0082]

As described above, according to the present invention, coded data is selected and multiplexed based on the amount of data stored in each multiplexing buffer means. Can flexibly respond to changes in the transmission speed of the media, and can suppress transmission delay.

Further, by multiplexing the data having a large amount of accumulated data, the required capacity of the multiplexing buffer means can be minimized and the delay time associated with the multiplexing can be reduced. .

By weighting the amount of stored data, encoded data having a small allowable delay time caused by multiplexing can be preferentially multiplexed, and the delay time can be minimized. be able to.

Further, the speed of the multiplexed output generated by inserting invalid data can be adjusted.

Further, the transmission control data can be periodically and preferentially multiplexed by the interrupt timer means irrespective of the amount of data stored in the multiplexing buffer means.

Further, by transmitting transmission control data instead of invalid data, multiplexing efficiency is improved.

The multiplexing buffer means is provided with a control signal output means for outputting a control signal of a generated code amount of the coded data based on the accumulated data amount, thereby controlling the coded data generation amount. Therefore, it is possible to prevent the accumulated amount from overflowing the capacity of the multiplexing buffer means, suppress the multiplexed buffer accumulated amount, and reduce the multiplexing delay.

Further, by providing the multiplexing control means with a control signal output means for outputting a control signal of the generated code amount of the coded data based on the accumulated data amount, the code amount can be reduced by the delay amount of the entire multiplexing apparatus. Since the amount of information generated by the multiplexing unit can be controlled, the multiplexing delay time of the entire multiplexing device can be reduced, and the multiplexing efficiency can be improved.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a multiplexing device according to an embodiment of the present invention.

FIG. 2 is a configuration diagram illustrating a multiplexing control unit of the multiplexing device according to the embodiment of the present invention.

FIG. 3 is an explanatory diagram showing an operation of the multiplexing device according to the embodiment of the present invention.

FIG. 4 is an explanatory diagram showing a configuration example of a fixed-length packet according to the embodiment of the present invention.

FIG. 5 is a configuration diagram showing a multiplex control unit of the multiplexing device according to the embodiment of the present invention.

FIG. 6 is a configuration diagram showing a part of the multiplexing device according to the embodiment of the present invention.

FIG. 7 is a configuration diagram showing a part of a multiplexing apparatus according to an embodiment of the present invention.

FIG. 8 is an explanatory diagram showing an operation of the multiplexing device according to the embodiment of the present invention.

FIG. 9 is a configuration diagram of a multiplexing device according to an embodiment of the present invention.

FIG. 10 is a configuration diagram of a multiplexing device according to an embodiment of the present invention.

FIG. 11 is a configuration diagram illustrating an example of a conventional multiplexer.

FIG. 12 is an explanatory diagram showing a configuration of a multiplexed frame in a conventional multiplexing device.

FIG. 13 is a configuration diagram showing another multiplexing device of a conventional example.

[Explanation of symbols]

 10a to 10n Encoding unit 11a to 11n Multiplexing buffer unit 12 Multiplexing unit 13 Multiplexing control unit 27 Buffer storage amount comparison unit 40 Transmission control data generation unit 41 Interrupt timer unit 42 Multiplexing unit 50 Weighting coefficient table 51 Weighting unit 52 Comparison unit 80 Encoding unit 81 Multiplex control unit

Continuation of front page (56) References JP-A-52-38811 (JP, A) JP-A-2-280439 (JP, A) JP-A-4-2199041 (JP, A) Transactions of the Institute of Electronics, Information and Communication Engineers, Vol . J 72-BI, No. 2, Yoshinori Sakai, Multimedia burst multiplexing method in communication terminals, pp. 146-64. 109-117 Proceedings of the IEICE Spring Conference, B-613 (1992-3-15), Tsukasa Okamoto et al., Examination of OAM cell insertion method for ATM layer quality monitoring, p3-180 (58) Field surveyed (Int. Cl. ⁷ , DB name) H04L 3/00-3/26 H04L 12/28

Claims (9)

(57) [Claims] 1. Encoded data in packet format to be input
And store the encoded data
The accumulated amount is grasped in bit units, and
A plurality of multiplexing buffer means for outputting an accumulated amount of coded data; and a code for each bit in each of the multiplexing buffer means.
A plurality of multiplexing buffers for transmitting the coded data packets to be multiplexed and output based on the amount of coded data stored data;
Multiplexing control means for selecting from packets of the encoded data stored in the means, and the multiplexing buffers based on the selection of the multiplexing control means.
The encoded data stored in the
A multiplexing device comprising: multiplexing means for reading out and multiplexing and outputting. 2. The multiplexing control means, wherein each of the multiplexing buffers is
The bit-unit coded data storage data in the
2. The multiplexing apparatus according to claim 1, wherein, among the data amounts, the multiplexed output is preferentially selected as the multiplexed output from the coded data stored in the multiplexing buffer means having the largest storage amount. 3. The multiplexing control means, wherein each of the multiplexing buffers is
The bit-unit coded data storage data in the
Weighting means for weighting the data amount , and the encoded data in bit units weighted by the weighting means.
2. The multiplexing apparatus according to claim 1, wherein the multiplexing output is selected based on the data accumulation amount . 4. An invalid data generating means for generating invalid data, wherein said multiplexing control means includes means for multiplexing each of said multiplexing buffers.
Means for storing the encoded data in bit units in the means
If neither the amount is less than the predetermined amount, the multiplexing device according to claim 1, wherein the selecting invalid data output of said invalid data generating means as a multiplexed output. 5. A transmission control data generating means for generating transmission control data, and an interrupt timer means for measuring an interval for transmitting the transmission control data, wherein the multiplex control means is based on a result of the time measurement by the interrupt timer means. 5. The multiplexing apparatus according to claim 1, wherein the transmission control data output from said transmission control data generating means is selected as a multiplexed output. 6. The multiplexing control means, wherein each of the multiplexing buffers is provided.
The bit-unit coded data storage data in the
When the data amount is less than a predetermined amount, the transmission control data output from the transmission control data generating means is selected as a multiplexed output, and the interrupt timer means is set to a time counting start state. The multiplexing device according to claim 5, wherein 7. The multiplexing buffer means includes a bit unit.
7. A control signal output means according to claim 1, further comprising control signal output means for outputting a control signal of a generated code amount of said encoded data to be inputted based on the encoded data accumulated data amount of the order. Multiplexer. 8. The multiplexing control means, wherein each of the multiplexing buffers is provided.
The bit-unit coded data storage data in the
Based on chromatography data amount, the multiplexing device according to 6 or claims 1, characterized in that a control signal output means for outputting a control signal of the generated code amount of the coded data input. 9. Encoding a plurality of multimedia data
Input multiple encoded data packets and store
In the multiplexing method of stacking and multiplexing, packets of the plurality of encoded data to be input are
In addition, when storing the plurality of encoded data,
The product volume is grasped in bit units, and this grasped
Based on the amount of coded data stored in each bit unit,
Code to be multiplexed and output from the encoded data packet
A multiplexing method comprising selecting and multiplexing packets of encoded data .